Friedreich's ataxia (FA) is a recessive, predominantly neurodegenerative disorder caused in most cases by mutations in the first intron of the frataxin (FXN) gene. This mutation drives the expansion of a homozygous GAA repeat that results in decreased levels of FXN transcription and frataxin protein. Frataxin (Fxn) is a ubiquitous mitochondrial protein involved in iron-sulfur cluster biogenesis, and a decrease in the levels of this protein is responsible for the symptoms observed in the disease. Although the pathological manifestations of FA are mainly observed in neurons of both the central and peripheral nervous system, it is not clear if changes in non-neuronal cells may also contribute to the pathogenesis of FA, as recently suggested for other neurodegenerative disorders. Therefore, the aims of this study were to generate and characterize a cell model of Fxn deficiency in human astrocytes (HAs) and to evaluate the possible involvement of non-cell autonomous processes in FA. To knockdown frataxin in vitro, we transduced HAs with a specific shRNA lentivirus (shRNA37), which produced a decrease in both frataxin mRNA and protein expression, along with mitochondrial superoxide production, and signs of p53-mediated cell cycle arrest and apoptotic cell death. To test for non-cell autonomous interactions we cultured wild-type mouse neurons in the presence of frataxin-deficient astrocyte conditioned medium, which provoked a delay in the maturation of these neurons, a decrease in neurite length and enhanced cell death. Our findings confirm a detrimental effect of frataxin silencing, not only for astrocytes, but also for neuron-glia interactions, underlining the need to take into account the role of non-cell autonomous processes in FA.